016c738188
Implement the xtask gen-schema binary at xtask/src/bin/gen_schema.rs that derives JSON Schema Draft 2020-12 from the Rust ExtractionResult type via the schemars crate. Changes: - Add stable key sorting (sort_keys_recursive) for byte-identical output - Set $id to stable URL: https://pdftract.com/schema/v1.0/pdftract.schema.json - Set title to "pdftract Output v1.0" - Add cargo alias `gen-schema` for convenient invocation - Emit schema to docs/schema/v1.0/pdftract.schema.json The schema is generated from the Rust types with schemars derives, ensuring the JSON schema is always in sync with the source types. Acceptance criteria: - cargo gen-schema regenerates docs/schema/v1.0/pdftract.schema.json - Generated schema validates against JSON Schema Draft 2020-12 - Schema $id is the stable URL - Title is "pdftract Output v1.0" - Stable ordering: regenerating twice produces byte-identical output - All expected types appear in $defs (BlockJson, SpanJson, PageResult, etc.) Note: page_type and confidence_source enums are not yet implemented in the Rust types (marked as TODO in schema/mod.rs). These will be added by sibling beads pdftract-1ob and pdftract-1f8we respectively. Closes: pdftract-5nv9h
27 lines
726 B
TOML
27 lines
726 B
TOML
# Cargo config for pdftract workspace
|
|
# Build aliases used by CI workflows and local development
|
|
|
|
[alias]
|
|
# CI-compatible aliases
|
|
bench-ci = "bench --features benchmark"
|
|
test-ci = "test --workspace"
|
|
|
|
# Development conveniences
|
|
b = "build"
|
|
br = "build --release"
|
|
c = "check"
|
|
cr = "check --release"
|
|
t = "test"
|
|
tr = "test --release"
|
|
|
|
# xtask aliases (invoke via --manifest-path to avoid workspace issues)
|
|
gen-schema = "run --manifest-path=xtask/Cargo.toml --bin gen_schema"
|
|
|
|
# Profile for CI property tests (nextest with proptest)
|
|
[profile.ci-proptest]
|
|
inherits = "release"
|
|
opt-level = 2 # Faster builds than full release, still fast execution
|
|
debug = false
|
|
strip = "none"
|
|
lto = "off"
|
|
codegen-units = 256 # Maximum parallelism
|